This is it !

[gmeast]

funny looking

I like the sound of what you're doing Greg. I thought it might need to get a
bit complicated before it could be simplified. Like you have done by the sound
of it.

I managed to get another primary on, thats two, no spacers or anything yet,
I pulsed them with a fet each, but in union. The waveform is a bit funny looking.

http://kobcsq.bay.livefilestore.com/...004.JPG?psid=1

http://kobcsq.bay.livefilestore.com/...006.JPG?psid=1

Cheers

Hi Farmhand,

That is NOT 'funny looking'. The 'waveform' is beginning to look like 'NOT a waveform'. The subtle differences between the two windings and their non-diametric placement on the core make them 'unrelated from a 'phase' standpoint. It looks 'fuzzy in the middle' because there might just be Non-Hertzian components 'peeking through'. A conventional core material would likely 'confine, prevent from forming, block' that energy. I think you are seeing a manifestation of LE there ... a 'crack', so to speak.

This looks GOOD!

Later

[Farmhand]

Hi Greg, I agree I think the shot is interesting. I made a small breakthrough
with decifering the behavior of this little CMOS circuit. It seems it's taylor
made in simplicity for the 200 - 500 uH coils. With a coil of 3 Henries things
change. A more sophisticated PWM like what you're working on won't care
about anything else and just work.

I think the CMOS setup I have been using is a simple relaxation oscillator.
Would that be correct? Thats why it's tied in with the circuit it's all part of
the oscillator kind of.

Anyway with some playing around with the tuning I can get 60 Khz with only
150 Ma and some dead time left over. At that frequency and narrow PW the
pulses are a lot less square.

This would be impossible to get anywhere without a scope. You guys were
right. Of course. Ive got such a mass of wire's and stuff I didn't remember to
get a shot, to busy trying to keep track of what i'm doing.

Hey Greg did you see the EPG of Stans I think it's fairly easy to see where
Steve Marks got his idea from. And why Steve Marks is not heard from.

This is all I can find out about it. It seems the patent has been hidden by the
scum that be. It's obviously a very usefull and powerfull tech.

Extract.

MEYER STANLEY A
AbstractAbstract of CA1213671 ELECTRICAL PARTICLE GENERATOR An electrical particle generator comprising a non-magnetic pipe in a closed loop having a substantial amount of magnetized particles encapsulated therein. A magnetic accelerator assembly positioned on said pipe having an inductive primary winding and a low voltage input to said winding. A secondary winding positioned on said pipe opposite to said primary winding. Upon voltage being applied to said primary winding said magnetized particles are passed through said magnetic accelerator assembly with increased velocity. The velocity accelerated particles continuing through said pipe induce an electrical voltage/current potential as they pass through the secondary winding. The increased secondary voltage is utilized in an amplifier arrangement

It's a little bit strange how stan didn't promote it's use for electrical power production. I guess he was transfixed by the water. I mean he had to perfect it. And he patented it. Stan had a lot of old school tech made modern.

Awesome work he did.

[7imix]

It's a little bit strange how stan didn't promote it's use for electrical power production. I guess he was transfixed by the water. I mean he had to perfect it. And he patented it. Stan had a lot of old school tech made modern.

Awesome work he did.

after watching a few of the Stan videos that have been linked to I really do understand where he was coming from. The water fuel cell implemented right in the spark plug and the laser gas pedal would be a very inexpensive retrofit for every existing internal combustion engine out there. The problem with electrical generators is that none of our cars use electric motors (yet) and so the retrofit is much more expensive, since the entire engine has to be replaced.

Too bad stan's faith in god wasn't enough to keep him alive... I wish he were still around.

[sucahyo]

Scope Shots.
http://kobcsq.bay.livefilestore.com/...001.JPG?psid=1

http://public.bay.livefilestore.com/...002.JPG?psid=1

Cool wave form, look like the wave form could use FWBR recovery instead of one diode.

Maybe similar to adding led in parallel on secondary but with one had inverted polarity.

P.S. My experiments tell me that thick aluminium shielding causes more output but also more input with helical coils. Toroid effect is still unknown to me.

Very interesting .

If someone knows for sure whether that FOR-NEXT loop is the frequency determinate or the pulse-width determinate, please pop in.

My guess is that run on 2 cycle of max controller speed. IF controller run at 6Mhz, then it would be 3Mhz. But then again, W mention he had faster code, meaning speed/frequency is not too important.


I think Stan pursue for patent which is the reason for why he don't use other people solution.

[7imix]

I went back and watched video 3 again to pay close attention to the assembly.

The first bunch of code before he zooms in and scrolls down is just housekeeping, setting up all the ports to be outputs and setting up some timers. What do the timers do? There are three of them, all set to 1024.

The bottom bit is essentially this, modified slightly for clarity:

Do
For Y = 0 to 6
Pulseout portd , Y , 1
Pulseout portb , Y , 1
Next
Loop

The Do ... Loop is an infinite loop. The For ... Next loop iterates over the six channels. He is using only ports 1, 3, and 5, but is pulsing all channels with the code. This basically halves the frequency. He mentions he gets even faster speeds (and presumably better results) by pulsing only the channels he is using, which basically removes the noops which is essentially what happens when he pulses an empty channel.

Note that he mentions making sure the board is set to use an external timing crystal, not the onboard clock for the chip.

It would be nice if I could find documentation for the language and environment he is using. What are the arguments to Pulseout? What is the distinction between portd and portb? The Y argument is obviously the channel to pulse. What is the D argument, which is always set to 1? The number of clock cycles to pulse the port?

I will look around and see if I can find documentation for the language and environment he is using.

[7imix]

Ok, the ide is bascom avr BASCOM-AVR - MCS Electronics

It's a dialect of basic that compiles down to machine code.

Pulseout toggles the state of a port back and forth (from the state it was originally in)

BASCOM-AVR

So we have "ports" and "pins". He pulses the same two ports every time through, but changes the pin number he is pulsing. Why pulse two ports? I don't get that part. The final parameter, which is always 1 in the video, is the duration of the pulse. From the above doc link

Number of periods the pulse will last. The periods are in uS when an XTAL of 4 MHz is used.

Since he is using a custom crystal, we don't know what the overall speed of the chip is. Perhaps the part list posted earlier will show it.

It doesn't seem like tuning the frequency is all that important for getting results. He seems to imply it works whether he pulses the unused channels or not. Pulsing the unused channels would give half the frequency that pulsing only used channels would give. I'm going to guess that tuning the frequency will improve the results but is not required to see modest results.

He does mention that when using the normal water fuel cell code the pulse width is 8 to 10% and it does not produce charging effects. With his stripped down code he gets pulse widths of 0.3 and very sharp rise times and fall times.

Whatever else may be required to get this working, extremely short pulses with extremely fast rise and fall times are going to be key.

[7imix]

Port d and port b docs.

BASCOM-AVR
BASCOM-AVR

I'm confused why pulsing one port instead of both isn't enough. Maybe it is.

Greg, the for loop and the external crystal used determines the frequency of pulses. The pulse width is always 1, which is basically the shortest possible pulse that combination of chip and crystal can provide. The frequency of pulses are basically as fast as the chip/crystal can go, with the exception of pulsing the unused mosfets in between the used ones, which he says is not required. So basically, the shortest pulse width at the highest frequency possible.

[Farmhand]

Port d and port b docs.

BASCOM-AVR
BASCOM-AVR

I'm confused why pulsing one port instead of both isn't enough. Maybe it is.

Greg, the for loop and the external crystal used determines the frequency of pulses. The pulse width is always 1, which is basically the shortest possible pulse that combination of chip and crystal can provide. The frequency of pulses are basically as fast as the chip/crystal can go, with the exception of pulsing the unused mosfets in between the used ones, which he says is not required. So basically, the shortest pulse width at the highest frequency possible.

The only thing I can think of real quick is the second port could be the rest
time or (dead time), since the pulses are so short and sharp, maybe he
needed to waste a bit of time or clock cycles. That just occured to me
because my PIC board came and i was looking at some sample code and it
used a method similar to that I think, it's hard for me to say I could try to
find that bit of the examples again.

It would be different code but the tutorial said that what that bit of code
was doing was wasting time.

Like one port is the "on" time and the other is the "off" time or extra "off" time.. Maybe

Cheers

[Farmhand]

Here's some code from a guy called Nigel. He references both port A and port
B, kinda makes a bit of sense to me. Sorry for posting random code. It just
caught my attention in my memory. And I thought it might help.

This simple program repeatedly switches all the output pins high and low.

;Tutorial 1.1 - Nigel Goodwin 2002
LIST p=16F628 ;tell assembler what chip we are using
include "P16F628.inc" ;include the defaults for the chip
__config 0x3D18 ;sets the configuration settings
;(oscillator type etc.)

org 0x0000 ;org sets the origin, 0x0000 for the 16F628,
;this is where the program starts running
movlw 0x07
movwf CMCON ;turn comparators off (make it like a 16F84)

bsf STATUS, RP0 ;select bank 1
movlw b'00000000' ;set PortB all outputs
movwf TRISB
movwf TRISA ;set PortA all outputs
bcf STATUS, RP0 ;select bank 0

Loop
movlw 0xff
movwf PORTA ;set all bits on
movwf PORTB
nop ;the nop's make up the time taken by the goto
nop ;giving a square wave output
movlw 0x00
movwf PORTA
movwf PORTB ;set all bits off
goto Loop ;go back and do it again

end

For some reason the quote bunchet it up.

[thatsthat]

Hex controller vs PWM3G board

@Farmhand and 7imix

I've been watching this thread for the past month and you guys are making good progress.

I have not posted anything on this forum, but I felt I can help with questions relating to AVR microprocessors specifically, as I have done development on it.

The PWM3G board does not contain any parts to support an Atmel AVR microprocessor. You need a small amount of circuit components to get it going. The crystal you mention 7imix, will not show up in the part list you referred to because it is not the Hex controller circuit.

The general speed of an AVR is 16 to 20MHz when using external oscillator circuit in the form of a crystal or resonator. Look at ATMega32 for detail on such a device.

The maximum internal clock speed is 8MHz for this micro. That means a pulse of only 0.125us (duty cycle of 0.6% of 48KHZ). At 20MHz the pulse shortens to only 0.050us (0.24% at 48KHZ). This second figure is pretty close to what watkykjy has mentioned. Don't get carried away though. If the MOSFET driver circuit can't switch this fast, you'll have trouble using such a short pulse, but I think there is more than enough brain power in this thread to solve that.

IMHO the AVR micro only makes control and timing of the trigger pulse to the MOSFETs a walk in the park but is NOT a requirement. This opinion have been raised by other members contributing to this thread. It is expensive, requires programming know how and special programming hardware to get a AVR going. It gets very easy to use once you have some things in place, but still.

I'll watch the videos specific to the BASCOM code and comment if I see anything specific.

Farmhand, your coil looks very good!

[thatsthat]

Got it

I examined video nr 3 at 5:07 and 5:11 and this is my observations.

The AVR is setup for 20MHz operation in the BASCOM code.

The line of code on the video says:
crystal = 20000000 ;use ATMega48 destinations

That means exactly what is says, 20million cycles/hertz.
Also, its a ATMega48.

The external crystal is a 20MHz.

Only the following pins are used as output on the micro.

PORTB: pin1,2,3
PORTD: pin2,3,4,5,6

Something else that might be unrelated to the pulsing, the Analog to Digital Converter is set up and started. This might be to measure voltage at a specific point in this circuit. It is never used from examining the code, as the code goes into an infinite loop just pulsing away.

Farmhand, what PIC is used on your development board?

[Farmhand]

G'day thatsthat, The board came with a PIC16F877A . Thanks a lot for your
help here, much appreciated. I hope you've enjoyed our "saga" as it seems to
be turning out to be. "This is It Thread" could be bigger than "Days of Our Live's"
one day.

Anyway i'm confident that even though I am almost at the limit of my
electronic skills and knowledge, thanks to the other guys and whoever else
can help, like you, we will get there.

Groovy name you got too.

Cheers.

Oh I almost forgot.

[sucahyo]

Do
For Y = 0 to 6
Pulseout portd , Y , 1
Pulseout portb , Y , 1
Next
Loop

The Do ... Loop is an infinite loop. The For ... Next loop iterates over the six channels. He is using only ports 1, 3, and 5, but is pulsing all channels with the code. This basically halves the frequency. He mentions he gets even faster speeds (and presumably better results) by pulsing only the channels he is using, which basically removes the noops which is essentially what happens when he pulses an empty channel.

Other important point for that code is he turn it on in sequence.
= pulse 1, pause, pulse 3, pause, pulse 5 pause, .... repeat


maybe faster code is
= pulse 1, pulse 3, pulse 5, .... repeat


Edit, above consider Y refer to channel number.....

[7imix]

Farmhand, you might be right, especially with some of the info thatsthat gave about the port configurations. The two Pulseout calls might just be generating the equivalent of a noop -- a delay.

Thatsthat, thanks a lot for chiming in. Very helpful. So are you saying there are parts on the assembled board that are not included in the parts list, that are just general support parts for the micro? That makes sense.

I think my conclusion about the pulsing code is that it is copy paste code. I don't think there are any huge secrets of the universe in there.

We definitely do not need a micro for this project, but a micro running at a high clock rate just happens to be a relatively easy way to get very short pulses at high frequency.

Driving the mosfets fully on and off with as short of a pulse as possible is going to be the tricky part. I have a nice and easy high frequency driver (a variation of the inverter circuit) running right now. It even drives through an optocoupler nicely. But when I try to bring the pulse width down the MOSFET doesn't shut off completely and gets hot. Next I plan on adding the npn/pnp circuit that was discussed a while back in this thread, which is designed to fill and drain the MOSFET gate capacitance as quickly as possible.

[thatsthat]

Greetings

@Farmhand,

Thank for the warm greeting.

I'm a digital electronics and programming guy. The analog electronics I'll leave to you guys, as it seems you got it sorted out pretty much.

Saga, what saga. It's more like David vs Goliath.

I leave the conspiracy to others to conspire around. I'm interested only in producing experimental results and evidence to prove claims others make and providing my "free" time to help others like you who refuse to accept the institutions view on physics and alternative energy solutions.

I'll be watching from a distance as I have no intention of building this "self" charging battery circuit for now.

I'm curious to see what the "analog" pulsing circuit looks like that removes the need for fancy micro controller etc.

All the best with your projects.

Cheers.